Invasion of tumor cells is a primary reason for the failure of surgical excision and focal radiation therapy to significantly improve the prognosis for patients with gliomas. No non-invasive method exists to detect the extent of invasion, thus the inability to target such areas with therapy. The overall goal of the proposed research is to develop and validate a diffusion-based MRI (magnetic resonance imaging) approach for the detection of glioma invasion. Our approach is to acquire MR signal using multiple diffusion-weightings (bvalues) and apply a stretched exponential analysis of the data, a new approach that we refer to as "alpha"DWI. Three metrics, the heterogeneity index ("alpha"), the distributed diffusion coefficient DDC, and the nth moments of the sub-voxel diffusion rates E(1/D^n), are derived. The stretched-exponential may be preferred to other models of diffusion because it allows one to estimate the sub-voxel rather than the inter-voxel distribution of water diffusion rates, and avoids making assumptions about the number of diffusion components. Preliminary results obtained in a rat brain tumor model demonstrate changes on ?DWI images, which correspond spatially to areas of tumor invasion as indicated by the presence of fluorescence-labeled tumor cells. Additional results suggest that the sensitivity of this technique to invading tumor cells may be optimized with the correct choice of diffusion timing parameters. Though promising, this technique will only prove feasible if it can provide information specific for invading tumor cells that is not confounded by the presence of other pathologic processes such as edema or radiation effects, both of which are commonly present in untreated and treated glioma patients. To test the feasibility of our approach, studies will be performed in rat brain tumor models, inoculated with stably expressing DSred-C6 glioma cells, and in patients with brain tumors. After feasibility is demonstrated, a future proposal will combine ?DWI and perfusion MRI methods, which are also being developed in our laboratory, for the comprehensive evaluation of brain tumor invasion, angiogenesis, growth and recurrence. This comprehensive approach to brain tumor imaging should dramatically improve the treatment and prognosis for patients with brain tumors. [unreadable] [unreadable] [unreadable]